INDEX
Explanations
proper nouns related to the name "Hard"
references to a person's last name
New Auto-Interp
Negative Logits
uality
-0.89
umbn
-0.78
entric
-0.71
ript
-0.68
Helsinki
-0.67
Schneider
-0.66
atory
-0.66
Mens
-0.65
utions
-0.64
Sut
-0.64
POSITIVE LOGITS
ball
0.98
icult
0.94
Hard
0.93
hard
0.92
entimes
0.90
ened
0.89
Hard
0.85
lett
0.85
ness
0.84
wood
0.82
Activations Density 0.011%